The introduction of ChatGPT has sparked enormous public interest in large language (deep-learning) models, which have been sophisticated enough to perform well on a variety of tasks. One way people are using these models is to construct diets. The prompts often include food restrictions that are an obligatory part of everyday life for millions of people worldwide. The aim of this study was to investigate the safety and accuracy of 56 diets, constructed for hypothetical individuals who are allergic to food allergens. Four levels, corresponding to the "baseline" ability of ChatGPT without prompting for specifics, as well as its ability to prepare appropriate diets when an individual has an adverse food reaction to two allergens or solicits a low-calorie diet, were defined. Findings from our study demonstrated that ChatGPT, although generally accurate, has the potential to produce harmful diets. More common errors involve inaccuracies in portions or calories of food, meals, or diets. We discuss here how the accuracy of large language models could be increased and the trade-offs involved. We propose that prompting for elimination diets can serve as one way to assess differences between such models.

Potato protein is recognized as one of the most valuable nonanimal proteins due to the high content of essential amino acids. So far, it has not been used in human nutrition on a large scale due to technological limitations regarding its acquisition. In this study, the protein fraction of potato juice was concentrated with the use of membrane separation. The obtained potato juice protein concentrate (PJPC) was characterized in terms of nutritional value and biological activity, and the amino acid composition, mineral content, and antioxidant properties were determined. Moreover, in vitro cytotoxic activity against cancer cells of the gastrointestinal tract was investigated. The results of the present study indicate that PJPC is an excellent source of lysine and threonine, while leucine is its limiting amino acid, with an amino acid score (AAS) of 65%. Moreover, PJPC contains substantial amounts of Fe, Mn, K, and Cu. As demonstrated experimentally, PJPC is also characterized by higher antioxidant potential than potato itself. Biological activity, however, is not limited to antioxidant activity alone. Cytotoxicity studies using a gastric cancer cell line (Hs 746T), a colon cancer cell line (HT-29), and human colon normal cells (CCD 841 CoN) proved that PJPC is characterized by selective activity against cancer cells. It can thus be concluded that the developed method of producing protein concentrate from potato juice affords a product with moderate nutritional value and interesting biological activity.

Cricket powder, described in the literature as a source of nutrients, can be a valuable ingredient to supplement deficiencies in various food products. Work continues on the implementation of cricket powder in products that are widely consumed. The aim of this study was to obtain gluten-free bread with a superior nutritional profile by means of insect powder addition. Gluten-free breads enriched with 2%, 6%, and 10% of cricket (Acheta domesticus) powder were formulated and extensively characterized. The nutritional value, as well as antioxidant and β-glucuronidase activities, were assessed after simulated in vitro digestion. Addition of cricket powder significantly increased the nutritional value, both in terms of the protein content (exceeding two-, four-, and seven-fold the reference bread (RB), respectively) and above all mineral compounds. The most significant changes were observed for Cu, P, and Zn. A significant increase in the content of polyphenolic compounds and antioxidant activity in the enriched bread was also demonstrated; moreover, both values additionally increased after the digestion process. The total polyphenolic compounds content increased about five-fold from RB to bread with 10% CP (BCP10), and respectively about three-fold after digestion. Similarly, the total antioxidant capacity before digestion increased about four-fold, and after digestion about six-fold. The use of CP also reduced the undesirable activity of β-glucuronidase by 65.9% (RB vs. BCP10) in the small intestine, down to 78.9% in the large intestine. The influence of bread on the intestinal microflora was also evaluated, and no inhibitory effect on the growth of microflora was demonstrated, both beneficial (Bifidobacterium and Lactobacillus) and pathogenic (Enterococcus and Escherichia coli). Our results underscore the benefits of using cricket powder to increase the nutritional value and biological activity of gluten-free food products.

The high-performance liquid chromatography (HPLC) method with photodiode array (PDA) and fluorescence (FL) detection for the simultaneous separation and determination of caffeine and water-soluble vitamins is described. The method is relatively rapid: 13 compounds can be analyzed within 30 min: caffeine; ascorbic acid (vitamin C); thiamine (vitamin B₁); riboflavin 5-phosphate (FMN, flavin mononucleotide) and riboflavin (vitamin B₂); nicotinic acid and nicotinamide (vitamin B₃); pantothenic acid (vitamin B₅); pyridoxal, pyridoxamine, and pyridoxine (vitamin B₆); folic acid (vitamin B₉); and cyanocobalamin (vitamin B₁₂). This HPLC method is designated for quantification of caffeine and water-soluble vitamins in liquid and tablet energy drinks as well as for determination of vitamins in vitamin supplements. The method was validated in terms of linearity, limit of detection (LOD) and quantification (LOQ), accuracy, and instrument precision. The LODs determined using HPLC-PDA ranged from 13 to 121 ng mL⁻¹ and using HPLC-FL were between 8 and 61 ng mL⁻¹. Intra- and interday instrument precisions for liquid energy drink, expressed as relative standard deviation (RSD), were less than 1.87 and 1.73 %, respectively, and less than 1.44 and 1.83 %, respectively, for tablet energy drink. The method could be useful in the quantitative analysis of caffeine and of water-soluble vitamins in energy drinks and vitamin supplements.